• Keith L Davis, Muralikrishna Tangirala, Juliana L Meyers, and Wenhui Wei.
• RTI Health Solutions, Research Triangle Park, NC 27709, USA. kldavis@rti.org
• Curr Med Res Opin. 2013 Sep 1; 29 (9): 1083-91.

ObjectiveTo evaluate outcomes in insulin-naive patients with type 2 diabetes mellitus (T2DM) who initiated insulin glargine or insulin detemir.MethodsRetrospective data were analyzed from the US General Electric Centricity electronic medical records (EMR) database from patients (≥18 years old) with T2DM initiating insulin glargine or detemir between January 1, 2006, and December 31, 2010. Included patients had EMR data for ≥6 months prior to (baseline) and ≥12 months after (follow-up) the index date (date of first insulin prescription), and at least one OAD and/or GLP-1 receptor agonist prescription order during baseline, but no previous insulin prescription. Patients were matched on baseline characteristics 5:1, insulin glargine to detemir, to ameliorate selection bias. Outcomes assessed were persistence with insulin therapy, glycemic control, hypoglycemia, body weight, and body mass index over follow-up.ResultsInsulin glargine and detemir groups were similar in terms of gender (51.0% and 51.5% female, P = 0.7356), age (57.8 and 57.4 years, P = 0.3368), A1C (9.4% and 9.4%, P = 0.6642), and body weight (101.9 kg and 102.4 kg, P = 0.4920) at baseline. During follow-up, patients initiating insulin glargine were more persistent (80.1% vs 67.8%, P < 0.0001) and had a greater change in A1C (-1.11% vs -0.96%, P = 0.0479). Percentage change in weight (0.91% and 0.65%, P = 0.2734) and hypoglycemia prevalence (3.6% vs 4.1%, P = 0.4338) were similar between groups.ConclusionsResults from this real-world EMR analysis suggest that among T2DM patients, initiating insulin treatment with insulin glargine may be associated with better treatment persistence and glycemic control, with similar prevalence of hypoglycemia and weight change, compared with initiating with insulin detemir. This study is limited by the retrospective nature of the data collection using EMRs and inability to confirm accuracy and completeness of data by secondary chart review.

25 keywords...

hide…